arXiv subjects

Abhirup Datta

Publications and source records attributed to Abhirup Datta.

At least 109 records · Page 6Linked to original sources

arXiv2022

Unveiling The Origin of Peculiar Diffuse Radio Emission in Abell 1351

Abell 1351 is a massive merging cluster that hosts a giant radio halo and a bright radio edge blended in the halo. In this paper, we present the first ever spectral analysis of this cluster using GMRT 610 MHz and VLA 1.4 GHz archival data and discuss the radio edge property. Using \textit{Chandra} data, we report the first tentative detection of shock front at the location of the edge in A1351 with discontinuities in both X-ray surface brightness and temperature. Our analysis strengthens the previous claim of the detected "edge" being a high luminosity radio relic. The radio relic has an integrated spectral index $α= -1.63 \pm 0.33$ and radio power $P_{1.4 \mathrm{\ GHz}} = 4.46 \pm 0.61\times10^{24}$ W $Hz^{-1}$ with a largest Linear Size (LLS) of 570 kpc. The radio spectral index map shows steepening in the shock downstream region. Our analysis favours the scenario where the Diffusive Shock Acceleration (DSA) of particles is responsible for the origin of the radio relic in the presence of strong magnetic field. We have also estimated the magnetic field at the relic location assuming equipartition condition.

astro-ph.HE↗

arXiv2022

On the Origin of Diffuse Radio Emission in Abell 85 -- Insights from new GMRT Observations

Extended, steep, and ultra-steep spectrum radio emission in a galaxy cluster is usually associated with recent mergers. Simulations show that radio phoenixes are aged radio galaxy lobes whose emission reactivates when a low Mach shock compresses it. A85 hosts a textbook example of a radio phoenix at about 320 kpc southwest of the cluster center. We present a new high resolution 325 MHz GMRT radio map illustrating this radio phoenix's complex and filamentary structure. The full extent of the radio structure is revealed for the first time from these radio images of A85. Using archival \textit{Chandra} X-ray observations, we applied an automated 2-D shock finder to the X-ray surface brightness and Adaptive Circular Binning (ACB) temperature maps which confirmed a bow shock at the location of the radio phoenix. We also compared the Mach number from the X-ray data with the radio-derived Mach number in the same region using multi-frequency radio observations and find that they are consistent within the 1$σ$ error level.

astro-ph.HE↗

arXiv2022

Deep uGMRT observations of the ELAIS-North 1 field: statistical properties of radio--infrared relations up to $z \sim$2

Comprehending the radio--infrared (IR) relations of the faint extragalactic radio sources is important for using radio emission as a tracer of star-formation in high redshift ($z$) star-forming galaxies (SFGs). Using deep uGMRT observations of the ELAIS-N1 field in the 0.3--0.5\,GHz range, we study the statistical properties of the radio--IR relations and the variation of the `$q$-parameter' up to $z=2$ after broadly classifying the faint sources as SFGs and AGN. We find the dust temperature ($\tdust$) to increase with $z$. This gives rise to $\qmir$, measured at $24\,\upmu$m, to increase with $z$ as the peak of IR emission shifts towards shorter wavelengths, resulting in the largest scatter among different measures of $q$-parameters. $\qfir$ measured at $70\,\upmu$m, and $q_{\rm TIR}$ using total-IR (TIR) emission are largely unaffected by $\tdust$. We observe strong, non-linear correlations between the radio luminosities at 0.4 and 1.4\,GHz with $70\,\upmu$m luminosity and TIR luminosity($\ltir$). To assess the possible role of the radio-continuum spectrum in making the relations non-linear, for the first time we study them at high $z$ using integrated radio luminosity ($\lrc$) in the range 0.1--2\,GHz. In SFGs, the $\lrc$--$\ltir$ relation remains non-linear with a slope of $1.07\pm0.02$, has a factor of 2 lower scatter compared to monochromatic radio luminosities, and $\qtirbol$ decreases with $z$ as $\qtirbol = (2.27 \pm 0.03)\,(1+z)^{-0.12 \pm 0.03}$. A redshift variation of $q$ is a natural consequence of non-linearity. We suggest that a redshift evolution of magnetic field strengths and/or cosmic ray acceleration efficiency in high-$z$ SFGs could give rise to non-linear radio--IR relations.

astro-ph.GA↗

arXiv2022

Study of the equatorial ionosphere using the Giant Metrewave Radio Telescope (GMRT) at sub-GHz frequencies

Radio interferometers, which are designed to observe astrophysical objects in the universe, can also be used to study the Earth's ionosphere. Radio interferometers like the Giant Metrewave Radio Telescope (GMRT) detect variations in ionospheric total electron content (TEC) on a much wider spatial scale at a relatively higher sensitivity than traditional ionospheric probes like the Global Navigation Satellite System (GNSS). The hybrid configuration of the GMRT (compact core and extended arms) and its geographical location make this interferometer an excellent candidate to explore the sensitive regions between the northern crest of the Equatorial Ionization Anomaly (EIA) and the magnetic equator. For this work, a bright radio source, 3C68.2, is observed from post-midnight to post-sunrise ($\sim$\,9 hours) to study the ionospheric activities at solar-minima. This study presents data reduction and processing techniques to measure differential TEC ($δ\rm{TEC}$) between the set of antennas with an accuracy of $1\times10^{-3}$ TECU. Furthermore, using these $δ\rm{TEC}$ measurements, we have demonstrated techniques to compute the TEC gradient over the full array and micro-scale variation in two-dimensional TEC gradient surface. These variations are well equipped to probe ionospheric plasma, especially during the night-time. Our study, for the first time, reports the capability of the GMRT to detect ionospheric activities. Our result validates, compared to previous studies with VLA, LOFAR and MWA, the ionosphere over the GMRT is more active, which is expected due to its location near the magnetic equator.

astro-ph.IM↗

arXiv2022

A Comparative Analysis To Deal With Missing Spectral Information Caused By RFI In Cosmological HI 21CM Observations

We investigate the effect of radio-frequency interference (RFI) excision in estimating the cosmological \hi\ 21 cm power spectrum. Flagging of RFI-contaminated channels results in a non-uniform sampling of the instrumental bandpass response. Hence, the Fourier transformation (FT) of visibilities from frequency to delay domain contaminates the higher foreground-free delay modes, and separating the spectrally fluctuating \hi\ signal from spectrally smooth foregrounds becomes challenging. We have done a comparative analysis between two algorithms, one-dimensional CLEAN and Least Square Spectral Analysis (LSSA), which have been used widely to solve this issue in the literature. We test these algorithms using the simulated SKA-1 low observations in the presence of different RFI flagging scenarios. We find that in the presence of random flagging of data, both algorithms perform well and can mitigate the foreground leakage issue. But, CLEAN fails to restrict the foreground leakage in the presence of periodic and periodic plus broadband RFI flagging and gives an extra bias to the estimated power spectrum. However, LSSA can restrict the foreground leakage for these RFI flagging scenarios and gives an unbiased estimate of the \hi\ 21 cm power spectrum. We have also applied these algorithms to the upgraded GMRT observation and found that both CLEAN and LSSA give consistent results in the presence of realistic random flagging scenarios for this observed data set. This comparative analysis demonstrates the effectiveness and robustness of these two algorithms in estimating the \hi\ 21 cm power spectrum from the data set affected by different RFI scenarios.

astro-ph.CO↗

arXiv2022

Extracting the 21-cm Power Spectrum and the reionization parameters from mock datasets using Artificial Neural Networks

Detection of the \hi~ 21-cm power spectrum is one of the key science drivers of several ongoing and upcoming low-frequency radio interferometers. However, the major challenge in such observations come from bright foregrounds, whose accurate removal or avoidance is key to the success of these experiments. In this work, we demonstrate the use of artificial neural networks (ANNs) to extract the \hi~ 21-cm power spectrum from synthetic datasets and extract the reionization parameters from the \hi~ 21-cm power spectrum. For the first time, using a suite of simulations, we present an ANN based framework capable of extracting the \hi~ signal power spectrum directly from the total observed sky power spectrum (which contains the 21-cm signal, along with the foregrounds and effects of the instrument). To achieve this, we have used a combination of two separate neural networks sequentially. As the first step, \texttt{ANN1} predicts the 21-cm power spectrum directly from foreground corrupted synthetic datasets. In the second step, \texttt{ANN2} predicts the reionization parameters from the predicted \hi~ power spectra from \texttt{ANN1}. Our ANN-based framework is trained at a redshift of $9.01$, and for \kk-modes in the range, $\rm{0.17<\kk<0.37~Mpc^{-1}}$. We have tested the network's performance with mock datasets that include foregrounds and are corrupted with thermal noise, corresponding to $1080$ hrs of observations of the \textsc{ska-1 low} and \textsc{hera}. Using our ANN framework, we are able to recover the \hi~ power spectra with an accuracy of $\approx95-99\%$ for the different test sets. For the predicted astrophysical parameters, we have achieved an accuracy of $\approx~81-90\%$ and $\approx~50-60\%$ for the test sets corrupted with thermal noise corresponding to the \textsc{ska-1 low} and \textsc{hera}, respectively.

astro-ph.CO↗

arXiv2021

Graphical Gaussian Process Models for Highly Multivariate Spatial Data

For multivariate spatial Gaussian process (GP) models, customary specifications of cross-covariance functions do not exploit relational inter-variable graphs to ensure process-level conditional independence among the variables. This is undesirable, especially for highly multivariate settings, where popular cross-covariance functions such as the multivariate Matérn suffer from a "curse of dimensionality" as the number of parameters and floating point operations scale up in quadratic and cubic order, respectively, in the number of variables. We propose a class of multivariate "Graphical Gaussian Processes" using a general construction called "stitching" that crafts cross-covariance functions from graphs and ensures process-level conditional independence among variables. For the Matérn family of functions, stitching yields a multivariate GP whose univariate components are Matérn GPs, and conforms to process-level conditional independence as specified by the graphical model. For highly multivariate settings and decomposable graphical models, stitching offers massive computational gains and parameter dimension reduction. We demonstrate the utility of the graphical Matérn GP to jointly model highly multivariate spatial data using simulation examples and an application to air-pollution modelling.

stat.ME↗

arXiv2021

Sparse nearest neighbor Cholesky matrices in spatial statistics

Gaussian Processes (GP) is a staple in the toolkit of a spatial statistician. Well-documented computing roadblocks in the analysis of large geospatial datasets using Gaussian Processes have now been successfully mitigated via several recent statistical innovations. Nearest Neighbor Gaussian Processes (NNGP) has emerged as one of the leading candidates for such massive-scale geospatial analysis owing to their empirical success. This article reviews the connection of NNGP to sparse Cholesky factors of the spatial precision (inverse-covariance) matrices. Focus of the review is on these sparse Cholesky matrices which are versatile and have recently found many diverse applications beyond the primary usage of NNGP for fast parameter estimation and prediction in the spatial (generalized) linear models. In particular, we discuss applications of sparse NNGP Cholesky matrices to address multifaceted computational issues in spatial bootstrapping, simulation of large-scale realizations of Gaussian random fields, and extensions to non-parametric mean function estimation of a Gaussian Process using Random Forests. We also review a sparse-Cholesky-based model for areal (geographically-aggregated) data that addresses interpretability issues of existing areal models. Finally, we highlight some yet-to-be-addressed issues of such sparse Cholesky approximations that warrants further research.

stat.ME↗

arXiv2021

Bayesian Functional Registration of fMRI Activation Maps

Functional magnetic resonance imaging (fMRI) has provided invaluable insight into our understanding of human behavior. However, large inter-individual differences in both brain anatomy and functional localization after anatomical alignment remain a major limitation in conducting group analyses and performing population-level inference. This paper addresses this problem by developing and validating a new computational technique for reducing misalignment across individuals in functional brain systems by spatially transforming each subject's functional data to a common reference map. Our proposed Bayesian functional registration approach allows us to assess differences in brain function across subjects and individual differences in activation topology. It combines intensity-based and feature-based information into an integrated framework and allows inference to be performed on the transformation via the posterior samples. We evaluate the method in a simulation study and apply it to data from a study of thermal pain. We find that the proposed approach provides increased sensitivity for group-level inference.

stat.AP↗

arXiv2021

On the detection of multiple shock fronts in A1914 using deep Chandra X-ray observations

Here, we report the new detection of three shock fronts using archival \textit{Chandra} X-ray observations in Abell 1914, which also hosts a radio halo, a radio phoenix, and a head-tail galaxy. In this study, we report the X-ray shock front at the position of the radio phoenix, which further strengthens the scenario that radio phoenix traces old plasma that gets lit up when compressed by shock passage. We further analyze the thermodynamic structure of the cluster in detail. We create temperature maps of A1914 using three different techniques, Adaptive Circular Binning (ACB), Weighted Voronoi Tessellations (WVT), and Contour binning (Contbin) method. These thermodynamic maps, along with the pseudo pressure and the pseudo entropy maps for the cluster, are evidence of disturbed morphology produced by multiple merger events. These merger events create cluster-wide turbulence, which may re-accelerate the relativistic particles and result in a radio halo within the cluster. Further, comparing X-ray and radio images reveals that the radio halo is contained within two X-ray shock fronts. Our analysis suggests that A1914 has both equatorial shock and axial shock within the cluster's ICM. We proposed a dual merging scenario based on the shock position and analysis of the thermodynamic maps obtained from the deeper \textit{Chandra} X-ray observations.

astro-ph.HE↗

arXiv2021

Random Forests for dependent data

Random forest (RF) is one of the most popular methods for estimating regression functions. The local nature of the RF algorithm, based on intra-node means and variances, is ideal when errors are i.i.d. For dependent error processes like time series and spatial settings where data in all the nodes will be correlated, operating locally ignores this dependence. Also, RF will involve resampling of correlated data, violating the principles of bootstrap. Theoretically, consistency of RF has been established for i.i.d. errors, but little is known about the case of dependent errors. We propose RF-GLS, a novel extension of RF for dependent error processes in the same way Generalized Least Squares (GLS) fundamentally extends Ordinary Least Squares (OLS) for linear models under dependence. The key to this extension is the equivalent representation of the local decision-making in a regression tree as a global OLS optimization which is then replaced with a GLS loss to create a GLS-style regression tree. This also synergistically addresses the resampling issue, as the use of GLS loss amounts to resampling uncorrelated contrasts (pre-whitened data) instead of the correlated data. For spatial settings, RF-GLS can be used in conjunction with Gaussian Process correlated errors to generate kriging predictions at new locations. RF becomes a special case of RF-GLS with an identity working covariance matrix. We establish consistency of RF-GLS under beta- (absolutely regular) mixing error processes and show that this general result subsumes important cases like autoregressive time series and spatial Matern Gaussian Processes. As a byproduct, we also establish consistency of RF for beta-mixing processes, which to our knowledge, is the first such result for RF under dependence. We empirically demonstrate the improvement achieved by RF-GLS over RF for both estimation and prediction under dependence.

stat.ML↗

arXiv2021

An X-ray and Radio Study of the Hubble Frontier Field Cluster Abell S1063

We present results from \textit{Chandra} X-ray observations and 325 MHz Giant Metrewave Radio Telescope (GMRT) observations of the massive and X-ray luminous cluster of galaxies Abell S1063. We report the detection of large-scale \lq\lq excess brightness\rq\rq\ in the residual \textit{Chandra} X-ray surface brightness map, which extends at least 2.7 Mpc towards the north-east from the center of the cluster. We also present a high fidelity X-ray flux and temperature map using \textit{Chandra} archival data of 122 ksec, which shows the disturbed morphology in the cluster. The residual flux map shows the first observational confirmation of the merging axis proposed by earlier simulation by \citet{Gomez2012AJ....144...79G}. The average temperature within $R_{500}$ is $11.7 \pm 0.56$ keV, which makes AS1063 one of the hottest clusters in the nearby Universe. The integrated radio flux density at 325 MHz is found to be $62.0\pm6.3$ mJy. The integrated spectrum of the radio halo follows a power-law with a spectral index $α=-1.43\pm 0.13$. The radio halo is found to be significantly under-luminous, which favored for both the hadronic as well as the turbulent re-acceleration mechanism for its origin.

astro-ph.HE↗

arXiv2021

spNNGP R package for Nearest Neighbor Gaussian Process models

This paper describes and illustrates functionality of the spNNGP R package. The package provides a suite of spatial regression models for Gaussian and non-Gaussian point-referenced outcomes that are spatially indexed. The package implements several Markov chain Monte Carlo (MCMC) and MCMC-free Nearest Neighbor Gaussian Process (NNGP) models for inference about large spatial data. Non-Gaussian outcomes are modeled using a NNGP Polya-Gamma latent variable. OpenMP parallelization options are provided to take advantage of multiprocessor systems. Package features are illustrated using simulated and real data sets.

stat.CO↗

arXiv2021

Generalized Bayes Quantification Learning under Dataset Shift

Quantification learning is the task of prevalence estimation for a test population using predictions from a classifier trained on a different population. Quantification methods assume that the sensitivities and specificities of the classifier are either perfect or transportable from the training to the test population. These assumptions are inappropriate in the presence of dataset shift, when the misclassification rates in the training population are not representative of those for the test population. Quantification under dataset shift has been addressed only for single-class (categorical) predictions and assuming perfect knowledge of the true labels on a small subset of the test population. We propose generalized Bayes quantification learning (GBQL) that uses the entire compositional predictions from probabilistic classifiers and allows for uncertainty in true class labels for the limited labeled test data. Instead of positing a full model, we use a model-free Bayesian estimating equation approach to compositional data based only on a first-moment assumption. The idea will be useful in Bayesian compositional data analysis in general as it is robust to different generating mechanisms for compositional data and includes categorical outputs as a special case. We show how our method yields existing quantification approaches as special cases. Extension to an ensemble GBQL that uses predictions from multiple classifiers yielding inference robust to inclusion of a poor classifier is discussed. We outline a fast and efficient Gibbs sampler using a rounding and coarsening approximation to the loss functions. We also establish posterior consistency, asymptotic normality and valid coverage of interval estimates from GBQL, as well as finite sample posterior concentration rate. Empirical performance of GBQL is demonstrated through simulations and analysis of real data with evident dataset shift.

stat.ME↗

arXiv2021

Hierarchical Multivariate Directed Acyclic Graph Auto-Regressive (MDAGAR) models for spatial diseases mapping

Disease mapping is an important statistical tool used by epidemiologists to assess geographic variation in disease rates and identify lurking environmental risk factors from spatial patterns. Such maps rely upon spatial models for regionally aggregated data, where neighboring regions tend to exhibit similar outcomes than those farther apart. We contribute to the literature on multivariate disease mapping, which deals with measurements on multiple (two or more) diseases in each region. We aim to disentangle associations among the multiple diseases from spatial autocorrelation in each disease. We develop Multivariate Directed Acyclic Graphical Autoregression (MDAGAR) models to accommodate spatial and inter-disease dependence. The hierarchical construction imparts flexibility and richness, interpretability of spatial autocorrelation and inter-disease relationships, and computational ease, but depends upon the order in which the cancers are modeled. To obviate this, we demonstrate how Bayesian model selection and averaging across orders are easily achieved using bridge sampling. We compare our method with a competitor using simulation studies and present an application to multiple cancer mapping using data from the Surveillance, Epidemiology, and End Results (SEER) Program.

stat.ME↗

arXiv2020

First multi-redshift limits on post-Epoch of Reionization (post-EoR) 21 cm signal from z = 1.96 - 3.58 using uGMRT

Measurement of fluctuations in diffuse HI 21 cm background radiation from the post-reionization epoch (z < 6) is a promising avenue to probe the large-scale structure of the Universe and understand the evolution of galaxies. We observe the European Large-Area ISO Survey-North 1 (ELAIS-N1) field at 300-500 MHz using the upgraded Giant Meterwave Radio Telescope (uGMRT) and employ the 'foreground avoidance' technique to estimate the HI 21 cm power spectrum in the redshift range z = 1.96-3.58. Given the possible systematics that may remain in the data, we find the most stringent upper limits on the spherically averaged 21 cm power spectra at k~1.0 Mpc$^{-1}$ are (58.87 mK)$^2$, (61.49 mK)$^2$, (60.89 mK)$^2$, (105.85 mK)$^2$ at z = 1.96,2.19,2.62 and 3.58, respectively. We use this to constrain the product of neutral HI mass density (Omega_HI) and HI bias (b_HI) to the underlying dark matter density field, [Omega_HI*b_HI], as 0.09,0.11,0.12,0.24 at z=1.96,2.19,2.62,3.58, respectively. To the best of our knowledge these are the first limits on the HI 21 cm power spectra at the redshift range z = 1.96 - 3.58 and would play a significant role to constrain the models of galaxy formation and evolution.

astro-ph.CO↗

arXiv2020

Using Artificial Neural Networks to extract the 21-cm Global Signal from the EDGES data

The redshifted 21-cm signal of neutral Hydrogen is a promising probe into the period of evolution of our Universe when the first stars were formed (Cosmic Dawn), to the period where the entire Universe changed its state from being completely neutral to completely ionized (Reionization). The most striking feature of this line of neutral Hydrogen is that it can be observed across an entire frequency range as a sky-averaged continuous signature, or its fluctuations can be measured using an interferometer. However, the 21-cm signal is very faint and is dominated by a much brighter Galactic and extra-galactic foregrounds, making it an observational challenge. We have used different physical models to simulate various realizations of the 21-cm Global signals, including an excess radio background to match the amplitude of the EDGES 21-cm signal. First, we have used an artificial neural network (ANN) to extract the astrophysical parameters from these simulated datasets. Then, mock observations were generated by adding a physically motivated foreground model and an ANN was used to extract the astrophysical parameters from such data. The $R^2$ score of our predictions from the mock-observations is in the range of 0.65-0.89. We have used this ANN to predict the signal parameters giving the EDGES data as the input. We find that the reconstructed signal closely mimics the amplitude of the reported detection. The recovered parameters can be used to infer the physical state of the gas at high redshifts.

astro-ph.CO↗

arXiv2020

A low-frequency radio halo survey of the South Pole Telescope SZ-selected clusters with the GMRT

The presence of non-thermal electrons and large scale magnetic fields in the intra-cluster medium (ICM) is known through the detection of mega-parsec (Mpc) scale diffuse radio synchrotron emission. Although a significant amount of progress in finding new diffuse radio sources has happened in the last decade, most of the investigation has been constrained towards massive low-redshift clusters. In this work, we explore clusters with redshift $z>0.3$ in search of diffuse radio emission, at 325 MHz with the Giant Metrewave Radio Telescope (GMRT). This campaign has resulted in the discovery of 2 new radio halos (SPT-CL J0013-4906 and SPT-CL J0304-4401) along with 2 other detections (SPT-CL J2031-4037 and SPT-CL J2248-4431), previously reported (at 325 MHz) in the literature. In addition, we detect a halo candidate in 1 cluster in our sample, and upper limits for halos are placed in 8 clusters where no diffuse emission is detected. In the $P_{1.4} - L_\mathrm{X}$ plane, the detected halos follow the observed correlation, whereas the upper limits lie above the correlation line, indicating the possibility of future detection with sensitive observations.

astro-ph.HE↗